Metabolic and Phenotypic Responses of Greenhouse ... - Cell Press

Molecular Plant

•

Volume 5

•

Number 2

•

Pages 401–417

•

March 2012

RESEARCH ARTICLE

Metabolic and Phenotypic Responses of Greenhouse-Grown Maize Hybrids to Experimentally Controlled Drought Stress Sandra Witta, Luis Galiciab, Jan Liseca, Jill Cairnsc, Axel Tiessend, Jose Luis Arause, Natalia Palacios-Rojasb and Alisdair R. Ferniea,1 a b c d e

Max Planck Institute of Molecular Plant Physiology, 14476 Potsdam-Golm, Germany CIMMYT, Apdo. Postal 6–641, 06600 Mexico, DF, Mexico CIMMYT Int., PO Box MP 163, Mount Pleasant, Harare, Zimbabwe Departamento de Ingenierıà CINVESTAV Unidad Irapuato, KM 9.8, Libramiento Norte, CP 36821 Irapuato, Guanajuato, Mexico University of Barcelona, Department of Vegetal Biology, Faculty of Biology, Av. Diagonal 645, 08028 Barcelona, Spain

ABSTRACT Adaptation to abiotic stresses like drought is an important acquirement of agriculturally relevant crops like maize. Development of enhanced drought tolerance in crops grown in climatic zones where drought is a very dominant stress factor therefore plays an essential role in plant breeding. Previous studies demonstrated that corn yield potential and enhanced stress tolerance are associated traits. In this study, we analyzed six different maize hybrids for their ability to deal with drought stress in a greenhouse experiment. We were able to combine data from morphophysiological parameters measured under well-watered conditions and under water restriction with metabolic data from different organs. These different organs possessed distinct metabolite compositions, with the leaf blade displaying the most considerable metabolome changes following water deficiency. Whilst we could show a general increase in metabolite levels under drought stress, including changes in amino acids, sugars, sugar alcohols, and intermediates of the TCA cycle, these changes were not differential between maize hybrids that had previously been designated based on field trial data as either drought-tolerant or susceptible. The fact that data described here resulted from a greenhouse experiment with rather different growth conditions compared to natural ones in the field may explain why tolerance groups could not be confirmed in this study. We were, however, able to highlight several metabolites that displayed conserved responses to drought as well as metabolites whose levels correlated well with certain physiological traits. Key words:

carbon metabolism; metabolomics; water relations.

INTRODUCTION Although improved adaptation to abiotic stress has long been a pursuit of breeders, it has been difficult to achieve, partially due to the fact that it is a quantitative trait controlled by many different genes (Lopes et al., 2011). Currently, one of the most critical abiotic stresses is water deficiency and it can be anticipated that climate change will exacerbate this problem in the future. C4 plants such as maize are often considered to have mastered the art of drought control, particularly since they can maintain photosynthesis when their stomata are closed (Lopes et al., 2011). Despite this fact, drought is currently estimated to cause average annual yield losses in the C4 plant maize the region of 17% in the tropics (Edmeades et al., 1989) whilst, for some regions of southern Africa, yield losses within a season can approach as much as 60% (Rosen and Scott, 1992). That said, this problem is by no means confined

to tropical maizes, with breeding programs of considerable scale being carried out to enhance grain potential and yield stability in stress-prone environments for temperate maizes grown in the USA (Bruce et al., 2002). Another attempt to produce more drought-tolerant maize is the development of drought-tolerant germplasm sources such as those created by CIMMYT’s breeding programs (Edmeades et al., 1996a, 1996b, 1999), which can then be crossed with maize germplasm that is well adapted to the target environment. 1 To whom correspondence should be addressed. E-mail [email protected], tel. +49 (0)331 5678211.

ª The Author 2011. Published by the Molecular Plant Shanghai Editorial Office in association with Oxford University Press on behalf of CSPB and IPPE, SIBS, CAS. doi: 10.1093/mp/ssr102, Advance Access publication 15 December 2011 Received 28 September 2011; accepted 13 November 2011

402

|

Witt et al.

d

Drought-Stressed Maize

Importantly, increased yield and yield stability of recently developed tropical and temperate maize genotypes have been related to increased abiotic stress tolerance and, in particular, to drought stress (Tollenaar and Wu, 1999; Ba¨nziger et al., 2002, Araus et al., 2002). Under drought, maize ovule abortion appears to be related to the flux of carbohydrates to the young ear around flowering and concurrent photosynthesis is required to maintain this above threshold levels (Zinselmeier et al., 1995). Drought has additionally been shown to reduce invertase activities in the ovaries, which would also likely result in a reduced flux of hexose sugars, altered hormone balance, and ovary abortion (Zinselmeier et al., 2000; Chourey et al., 2010). Despite the above hints towards mechanisms that may be responsible for yield loss under drought and despite the fact that the molecular responses of a wide range of plants to drought stress have been well characterized (for reviews, see Bartels and Sunkar, 2005; Bohnert et al., 2006; Tardieu et al., 2011), the multigenic nature of drought tolerance has rendered application of this knowledge to crop improvement a considerable challenge. That said, considerable improvements in yield under drought selection have been made in both tropical and temperate maizes (for reviews, see Bruce et al., 2002; Lopes et al., 2011). Furthermore, the potential of marker-assisted breeding approaches is highly recognized and efforts have been made in the development of newer germplasm with improved stress tolerance (see Hoisington et al., 1996; Quarrie et al., 1999; Xu et al., 2009; Crossa et al., 2010) as are transgenic approaches exploiting modifications of target genes (Bartels and Nelson, 1994; Leung and Giraudat, 1998; Rontein et al., 2002). More recently, high throughput methods such as genome map-based and transcriptomic and metabolomic approaches that additionally highlight less obvious target genes have been employed as screening tools for target gene selection (Seki et al., 2001; Sun et al., 2001; Schauer et al., 2006; Harrigan et al., 2007a, 2007b; Semel et al., 2007). The combination of such methods will likely rapidly reduced the tens of thousands of possible genes to a handful of candidate genes for direct testing and will thus likely accelerate the target selection process (Fernie and Schauer, 2009). In this paper, we determined a range of morphophysiological parameters including plant height, ear height, leaf temperature, leaf stomatal conductance, and leaf chlorophyll content, whilst the metabolite composition was determined by gas-chromatography mass spectrometry (GC–MS) for leaf blades, ears, husks, sheath, and silks of six hybrids of the CIMMYT maize physiology program, in response to drought imposed under a controlled experimental greenhouse environment. The six CIMMYT hybrids selected have previously been designated as drought-susceptible, moderately tolerant, and tolerant to drought under field conditions (Edmeades et al., 1996a, 1996b; Elings et al., 1996; Monneveux et al., 2005, 2008; Cabrera-Bosquet et al., 2011; Pandey and Gardner 1992). This represents a highly comprehensive initial survey,

providing a more detailed inter-tissue analysis of a broad range of metabolites than carried out to date and providing important insight into the utility of these various tissues for biomarker identification. The obtained data were additionally carefully statistically evaluated with respect to whether plants could be discriminated on their genotype or the environmental conditions that they were subjected to. Finally, correlation analysis between metabolite and physiological traits was carried out in an attempt to identify relationships between these traits that may be of interest from a breeding perspective. The combined results are discussed in the context of the use of metabolomics as a tool to aid strategies for crop improvement.

RESULTS Description of the Experimental Set-Up In order to characterize the impact of water deficit on plant development, six different maize traits were tested related to their response after application of drought. The hybrid material used in this experiment was selected on the basis on field data collected under well-watered and drought-stress conditions in Tlaltizapan (Morelos, MX) in 2009. Grain yield, the anthesis-to-silking interval (ASI), ear aspect, and phenology were the parameters chosen to discriminate the genotypes. With respect to this dataset, three hybrids tolerant to drought and three hybrids displaying varying susceptibility to drought were used (Table 1). For this particular study, we were interested in analyzing the metabolic composition of five different tissues and identify the one with more metabolic information regarding drought stress. Plant height, leaf temperature, stomatal conductance, ear height, and chlorophyll content were taken in order to monitor the stress under greenhouse conditions. Six individuals per genotype were planted in the greenhouse (one plant per pot) and the water was controlled by drip irrigation. Two weeks before flowering, stress was applied by stopping irrigation and, following 2 weeks of stress (or, in the case of the control, continued watering), tissues were harvested and snap-frozen in liquid nitrogen for metabolite analysis.

Influence of Drought Stress on Morphophysiological Parameters in the Different Maize Hybrids The intensity of drought stress and its related influence on the development of the tested hybrids was investigated by measuring the abovementioned parameters. Prior to the application of stress, plant height was invariant both between the pre-stress and control samples of each genotype and between genotypes (Figure 1A). A second measurement after 13 d of stress application, however, revealed significantly lower plant height under drought for all tested genotypes (Figure 1B) together with genotypic differences in the response to drought (Figure 1C and 1D). However, there was surprisingly no effect of tolerance type on this parameter (Table 2). We next took measurements of leaf temperature in order to monitor the plant during stress. Prior to the stress application, no changes were observed (Figure 2A and 2C). However,

Witt et al.

d


|

403

Table 1. Genotypes Used in the Greenhouse Experiment. Pedigree

Origin

Tolerance to drought

NPE1

CML-486-B-B/CML-312 SR

Mexico, subtropical

Moderate

NPE2

CML311/MBR C2 Bc F41-2-BBBBB-B-B-B-B/CML-312 SR

Mexico, insect resistance

Susceptible

NPE3

[GQL5/[GQL5/[MSRXPOOL9]C1F2-205-1(OSU23i)-5–3X-X-1-BB]F2-4sx]-11–3–1–1-B*5-B-B/CML-312 SR

Zimbabwe, subtropical

Tolerant

NPE4

DTPWC9-F31-1–3–1–1-B-B-B-B-B/CML-312 SR

Mexico, subtropical

Tolerant

NPE5

CML311/MBR C3 Bc F95-2–2–1-B-B-B-B-B-B-B/CML-312 SR

Mexico, insect resistance

Moderate

NPE6

La Posta Seq C7-F96-1–2–1–3-B-B-B-B/CML-312 SR

Mexico, tropical

Tolerant

Source material is described in Elings et al. (1996), Edmeades et al. (1996a, 1996b), Cabrera-Bosquet et al. (2011), and Monneveux et al. (2005).

Figure 1. A Comparison of Drought Effect on Plant Height. (A) Black bars represent drought-stressed plants, gray bars well-watered plants. Plant height was measured before drought-stress application. (B) A second measurement of plant height was taken after 19 d of stress (or lack thereof). Each value represents the mean of six individual biological replicates. Asterisks represent values that were determined to be significantly different from the control using the Student’s ttest, P , 0.01. (C, D) Different treatments, genotypes, and drought tolerance are color-coded. Blue colors represent well-watered control plants; red colors show drought-stress-treated plants (from the left to the right: light blue/orange: susceptible genotypes, orange/blue: moderate genotypes, dark blue/orange: tolerant genotypes). Box plots of individual measurements are shown. (C) corresponds to data presented in (A); (D) corresponds to data presented in (B).

404

|

Witt et al.

d


Table 2. Two-Way ANOVA of the Physiological Data. Treatment

Tolerance

Interaction

Plant height_1

6.09E-01

3.49E-01

8.17E-01

Plant height_2

7.12E-11

5.89E-01

4.82E-01

Ear height

1.49E-02

7.89E-01

6.19E-01

Stomatal conductance_1

5.88E-15

6.62E-01

6.51E-01


5.58E-04

9.89E-01

3.51E-01

Leaf temperature_1

9.57E-01

7.18E-01

8.48E-01

Leaf temperature_7

7.17E-29

4.33E-02

8.57E-01

Chlorophyll content

1.46E-07

2.40E-03

5.13E-01

Treatment

Genotype

Interaction

Plant height_1

5.48E-01

1.45E-04

3.70E-01

Plant height_2

2.09E-18

9.33E-13

2.14E-01

Ear height

3.13E-05

1.09E-15

6.80E-01


5.83E-14

6.63E-01

8.72E-01


8.10E-04

6.08E-01

8.84E-01

Leaf temperature_1

9.58E-01

6.86E-01

7.44E-01

Leaf temperature_7

2.88E-27

1.71E-01

3.47E-01

Chlorophyll content

1.09E-08

6.78E-05

2.07E-01

The upper table represents a tolerance by treatment comparison; the lower table represents a treatment by genotype comparison. For both, all physiological data were used. Gray squares represent significant association (P , 0.01).

following 7 d of stress, clear changes as strong and highly significant increases in leaf temperature were observed in all hybrids in comparison to their corresponding control plants (Figure 2B and 2D). We next determined leaf stomatal conductance. The influence of drought stress on this trait was very rapid. Following 1 d of stress application, the stomatal conduction in stressed plants decreases rapidly in all hybrids (Figure 3A and 3C). However, following 7 d of stress, only NPE1 exhibits a slight but significant decrease in the stomatal conduction (Figure 3B and 3D) but no genotypic differences were recorded (Table 2). As a consequence, there is no statistical evidence for differences in the behavior of this trait between hybrids (Table 2). Ear height was significantly different following stress in NPE1, NPE2, NPE3, and NPE5, but not for NPE4 or NPE6 (Figure 4A and 4C), while genotypic differences were also present (Table 2). Chlorophyll content was decreased following drought stress for NPE1, NPE2, NPE3, NPE5, and NPE6, but not for NPE4 (Figure 4B and 4D), and genotypic differences were also significant (Table 2). Ear height was also significantly affected by both treatment and genotype (Table 2). In spite of genotypic differences existing for plant and ear height and chlorophyll content, differences between the subset of tolerant and susceptible genotypes were only present for the chlorophyll content (Table 2).

Metabolite Abundance Differs in Tested Tissues One of the aims in this study was to investigate relative metabolite abundance in different tissues of maize. The tested organs of leaf blade, leaf sheath, ear, husk, and silks differed

very strongly in their metabolite abundance, with blade tissue being the most diverse in terms of metabolites while containing considerably lower metabolite levels than the other tissues (Figure 5). A similar, although less strong, pattern was also seen in leaf sheath and silks; however, ears and husks differed greatly. In the ear, there was a considerable accumulation of nicotinamide, a water-soluble vitamin and part of the vitamin B complex and the organic nitrogen-containing compound urea.

Metabolic Profiling Allows Clear Separation of Tissue Type Subjecting the metabolite data to a principal component analysis (PCA) separates all tested tissue regarding to metabolite composition, independently from genotype or treatment situation (Figure 6A), reflecting that major differences exist in metabolite composition within the different tissues. PC1 reveals that blade tissue shows the most contrasting metabolic profile in comparison to other tissues. PC2 separates different treatments. A tendency for a treatment effect on the metabolic level is in fact visible for all tested tissues but, for blade tissue, this effect is most highly significant (Figure 6B). It is worth mentioning that, under well-watered conditions, different tested hybrids show a very similar metabolic pattern whereas, under drought, these hybrids show a more diverse metabolic composition with regard to their genetic background. Furthermore, the PCA also fails to resolve hybrids that were predicted to be stress-susceptible or apparently droughtstress-tolerant. Using the statistical tool ANOVA (Figure 6B), it is clear that the factor that harbors the most influence is tissue, followed by treatment, then genotype. We next assessed the metabolites in a point-by-point manner (the full dataset is available as Supplemental Figure 1), whereas selected metabolites associated with tissue type, drought treatment, or tolerance class are presented in Figures 7–9, respectively. Starting with those that were strongly related to tissue type, the metabolites xylose—a minor cell wall sugar, vanillic acid, and methionine were most closely associated with this feature—in all instances being at relatively low levels in the leaf blade (Figure 7). Looking at those metabolites varying with the drought treatment, perhaps not surprisingly, putrescine and proline, but also histidine, were amongst those prominently associated (Figure 8). By contrast, very few metabolites were closely associated with the tolerance class; however, beta-alanine and phosphoric acid were exceptional in this manner, particularly in the instance of the droughtstress-susceptible genotype NPE2 (Figure 9).

Several Drought-Stress-Related Changes Can Be Observed in the Metabolic Constitution of the Blade Tissue The PCA analysis described above suggested that blade tissue was the best to visualize metabolomic changes following drought stress. We next attempted to perform a point-bypoint analysis across all tissues. For this purpose, we present the data in a heatmap (Figure 10), in which values are normalized to their respective well-watered controls. The data of the heatmap reflect that of the PCA analysis in demonstrating that

Witt et al.

d


|

405

Figure 2. A Comparison of Drought Effect on Maize Leaf Temperature. Black bars represent drought-stressed plants, gray bars well-watered plants. (A) Maize leaf temperature was measured before drought-stress application. (B) A second measurement of maize leaf temperature was taken 7 d following stress application. Each value represents the mean of six individual biological replicates. Asterisks represent values that were determined to be significantly different from the control using the Student’s t-test, P , 0.01. (C, D) Different treatments, genotypes, and drought tolerance are color-coded. Blue colors represent well-watered control plants; red colors show drought-stress-treated plants (from the left to the right: light blue/orange: susceptible genotypes, orange/blue: moderate genotypes, dark blue/orange: tolerant genotypes). Box plots of individual measurements are shown. (C) corresponds to data presented in (A); (D) corresponds to data presented in (B).

leaf blade material definitely exhibits large metabolic changes following drought stress. For example, there were dramatic increases in tryptophan, phenylalanine, and histidine as well as in proline. By contrast, the levels of pyruvic acid decreased (in four of the six hybrids), as did the levels of quinic acid (in five of six hybrids) following drought stress. Shown heat map data also make clear that leaf sheath shows similar changes on metabolic level compared to leaf blade tissue due to the drought-stress response, but not so strong by far. Nevertheless, an increase in several amino acids like proline, isoleucine, and phenylalanine is visible and this is seen also for silks tissue. Slight increases directing especially sugar substances could be observed in ear tissue whereas, for husk tissue, hardly any consistent changes through all tested genotypes were found. These

data strongly support the contention from the PCA that leaf blade is the best tissue to study metabolic responses to drought stress, followed by leaf sheath and silk, but that ear and husk are relatively metabolically inert in response to water deficit. Evaluation of this entire dataset by ANOVA analysis revealed an impressive number of metabolites responding to drought across all genotypes with far fewer genotype-specific effects and fewer still tolerance group-specific effects on metabolite levels (Table 3).

Correlation of Specific Metabolites with Leaf Temperature and Stomatal Conductance An interesting aspect of this study was to combine knowledge from physiological response of maize following drought stress

406

|

Witt et al.

d


Figure 3. A Comparison of Drought Effect on Stomatal Conductance. Black bars represent drought-stressed plants, gray bars well-watered plants. (A) Stomatal conductance was measured 1 d following drought-stress application. (B) A second measurement of stomatal conductance after 7 d following drought-stress application. Each value represents the mean of six individual biological replicates. Asterisks represent values that were determined to be significantly different from the control using the Student’s t-test, P , 0.01. (C, D) Different treatments, genotypes, and drought tolerance are color-coded. Blue colors represent well-watered control plants; red colors show drought-stress-treated plants (from the left to the right: light blue/orange: susceptible genotypes, orange/blue: moderate genotypes, dark blue/orange: tolerant genotypes). Box plots of individual measurements are shown. (C) corresponds to data presented in (A); (D) corresponds to data presented in (B).

with metabolic data. For this purpose, a correlation analysis of the physiological parameters leaf temperature and leaf stomatal conductance and metabolic response was performed and results are presented in Figure 11. A strong negative correlation for the TCA cycle intermediates pyruvic acid and succinic acid with leaf temperature (and, by inference, water stress) was observed. Interestingly, the opposite is seen for the amino acids proline and isoleucine (Figure 11A). A positive correlation between stomatal conductance and pyruvic acid and succinic acid was also observed using the stomatal conductance data taken 1 d after application of drought stress (Figure 11B). Pyruvic acid positively correlated with stomatal conductance 1 d after application of drought stress, whilst the levels of phenylalanine, adenine, and putrescine negatively correlated with

these data (Figure 11C). It is important to note that this is only a subset of the analyses we performed and data for all significant correlations in all tested tissues are available in Supplemental Figures 2–5.

DISCUSSION The work presented here provides an initial controlled environment assessment of tropical maize hybrid response to drought stress at the level of the metabolome. Whilst considerable work has been carried out in other crop species, most notably tomato (for reviews, see Fernie et al., 2006; Fernie and Klee, 2011), only a handful of studies have been carried out using metabolomics in maize to date (Harrigan et al., 2007a,

Witt et al.

d


|

407

Figure 4. Influence of Drought on Ear Height and Chlorophyll Content. Black bars represent drought-stressed plants; gray bars well-watered plants. (A) Ear height under drought-stress conditions. Each value represents the mean of six individual biological replicates. Asterisks represent values that were determined to be significantly different from the control using the Student’s t-test, P , 0.01. (B) SPAD measurement of chlorophyll content. Each value represents the mean of six individual biological replicates. Asterisks represent values that were determined to be significantly different from the control using the Student’s t-test, P , 0.01. (C, D) Different treatments, genotypes, and drought tolerance are color-coded. Blue colors represent well-watered control plants; red colors show drought-stress-treated plants (from the left to the right: light blue/orange: susceptible genotypes, orange/blue: moderate genotypes, dark blue/orange: tolerant genotypes). Box plots of individual measurements are shown. (C) corresponds to data presented in (A); (D) corresponds to data presented in (B).

2007b; Lisec et al., 2011). This is, at first sight, surprising, given that an incredible research effort has been expended on improving metabolic traits. In maize, examples of this include the Illinois long-term selection program for oil and protein content (Moose et al., 2004) and the more recent association mapping approaches that revealed the enzymes responsible for kernel oil content and composition (Zheng et al., 2008) and pro-vitamin A content (Harjes et al., 2008). In conjunction with drought stress, many studies have been carried out aimed at understanding, for example, the genetic factors underlying

abscisic acid levels (Setter et al., 2011), osmolyte accumulation (Yancey, 2005), and the relationships between yield, ash content, and isotope distributions (Cabrera-Bosquet et al., 2009, 2011). Four previous studies have looked at a broader range of metabolites in maize breeding material (Harrigan et al., 2007a, 2007b; Ro¨misch-Margl et al., 2010; Lisec et al., 2011). The first of those of Harrigan and co-workers studied sugar, oil, amino acid, and organic acid content across a broad range of hybrids derived from 48 inbreds crossed against two different testers and grown under multiple environments and

408

|

Witt et al.

d


revealed that the variance in these compounds was considerable. Indeed, it exceeded that observed for analytes tested by the Organisation for Economic Co-operation and Development (OCED) for the introduction of new biotech crops (Harrigan et al., 2007b). Given the high variability inherent in the primary metabolome, these authors next evaluated the mean levels of the same metabolites as well as glycine, betaine, and absicic acid (Harrigan et al., 2007a). The other two studies were more concerned with heterosis—the superior performance of hybrids over their parents. The study of Ro¨misch-Margl et al. studies sugar and amino acid components in developing maize kernels (Ro¨misch-Margl et al., 2010), whilst that of Lisec and co-workers determined the levels of 112 metabolites in roots of corn hybrids. In contrast to previous studies in tomato fruits (Schauer et al., 2008), both groups found a considerable proportion of heterosis in these maize populations; however, neither study was repeated under adverse growth conditions. In contrast to the Harrigan et al. (2007a, 2007b) studies, we performed our experiments under controlled-environment greenhouse conditions and we evaluated metabolite leaf blades, ears, husks, sheath, and silks rather than merely the grains. In addition, it is important to note that the genotypes were tropical or subtropical as opposed to temperate. The selected genotypes in this study were defined as either tolerant, moderate, or susceptible on the basis of multiple field trials. That said, in the greenhouse, they did not display consistent difference from one another on the basis of these classifications, making it difficult to associate phenotypic characteristics of the plants with their supposed response to drought stress (see Figures 1–4). This conclusion was supported by the performance of two-way ANOVA tests across the physiological parameters that revealed that, whilst the genotypes were significantly different on an individual basis, they did not follow similar behavior in the greenhouse on the basis of their classifications. This was initially somewhat disappointing. But, the fact that a greenhouse trial differs in several aspects to a field trial may explain this phenomenon. Important environmental factors like light intensity, length of photoperiod per day, water status, or temperature differ in a natural setting compared to a strongly controlled greenhouse environment. Therefore, previous field trials were performed in the winter season whereas the greenhouse trial was carried out in the main season. In fact, this enlarges differences in growth conditions even more. However, it is clear from the dense dataset collected that several important observations could be made on the metabolic response to drought stress and as to which tissue would be the best to harvest diagnostic markers from. Such information is likely to be of high value for maize breeders, since it may aid in the generation of material that is more tolerant to drought. Worthy of mention in this vein is the proposal that has been muted that selection

Figure 5. Heatmap of Metabolite Levels in Different Tissues.

Red colors represent high metabolite levels; blue colors highlight low levels using a false-color scale. Values represent the mean of all genotypes and conditions (n = 72).

Witt et al.

A

B

Number of metabolites

PC2 (13.82%)

5

0

-5

Blade

-10 -10

Ear

-5

Husk

0

Sheet

Silks

5

PC1 (47.37%)

Tissue

d


Treatment 80

80

60

60

60

40

40

40

20

20

20

0 0.0 0.2 0.4 0.6 0.8 1.0

0 0.0 0.2 0.4 0.6 0.8 1.0

0.0 0.2 0.4 0.6 0.8 1.0

50

50

50

20

20

20

10

10

10

5

5

5

2

2

2

1

1

1

-200

10

-100

10

0

-200

10

409

Genotype

80

0

|

10

-100

10

0

10

-200

10

-100

10

0

10

ANOVA P-values (Bonferroni)

Figure 6. Principal Component Analysis (PCA) of Metabolite Profiles in Different Tissues. Blue color shows well-watered traits; red color represents drought-stressed traits. Bright colors highlight stress-susceptible traits; dark colors accent tolerant traits. Mean values and standard deviation are shown. (A) PCA shows clearly separation of metabolite composition within different tissues. Blade tissue can be pointed out here as the tissue that is separating the most from the other tissues. (B) Bonferroni-corrected ANOVA displays number of metabolites in which highly significant changes were observed regarding tissue, treatment, and genotype.

under drought may not be the most efficient approach for obtaining drought-tolerant crops (Bolanos and Edmeades, 1996). To address the question regarding the best tissue to use for understanding drought stress from a metabolic perspective, we performed both a PCA analysis (Figure 6) and a pointby-point metabolite analysis (Figure 10). These data revealed that the leaf blade clearly displayed the greatest metabolic response to drought stress and thus that, despite containing a fairly specific metabolome complement under non-stressed conditions (Figure 5), it is likely the best tissue to use for diagnostic studies of improving drought tolerance by metabolic mechanisms. Surveying the other tissues suggests that, whilst leaf sheath, leaf blade, and silks may be of some use in this respect, the relative metabolic inertness of ears and husks renders them inappropriate for this purpose. Among the plant parts analyzed, leaf blades are the most adequate to record changes in growing water conditions experienced by the plant. While husks and leaf sheaths are organs that, like the blades, remain active in the plant for a long time, their anatomical (xeromorphic) adaptations as well as their placement as a shell coating other parts of the plant expose them less to water stress than the blades (Ristic and Cass, 1981; Araus et al., 1993; Tambussi et al., 2007). Concerning the ears and silks, these are shortduration organs (which remain active for one or a few weeks) that are only active if fully hydrated and so do not register the changes in growing conditions. Thus, drought may dramatically shorten the duration of ears and silks but not their relative water content while these organs remain active.

A second major aim was to identify whether there were metabolic features that rendered certain hybrids susceptible and others resistant. Whilst we were not able to find strong statistical support for many such features, some interesting observations were apparent in our study. For example, in the leaf blades following drought stress, it is notable that amino acids increase in all but one of the hybrids—NPE1, which is the only fully susceptible genotype. Moreover, correlation analysis revealed strong positive correlation with proline with temperature and negative with stomatal conductance. It has been long documented that proline accumulates in response to drought and, for that, proline has been frequently argued to play a role in drought tolerance (Verbruggen and Hermans, 2008; Yang et al., 2010; Sharma et al., 2011) as well as a number of other amino acids such as isoleucine (see also Nunes-Nesi et al., 2010; Bowne et al., 2012; Kochevenko et al., 2012). In addition, increases in putrescine, pyruvic, and succinic acids in response to drought were highly consistent and, since polyamines have previously been implicated in drought stress (Alcazar et al., 2011), they may prove to be interesting targets in the future. Intrudingly, histidine was one of the amino acids that were elevated following drought stress in the study of Harrigan and co-workers (2007a). That these changes were consistent across both tropical and temperate maizes grown under dramatically different environmental conditions suggests that genetic manipulation of the enzymes responsible for their accumulation may represent an important strategy for future breeding of drought-tolerant maize.

410

|

Witt et al.

d


Figure 7. Box Plots of Selected Metabolite Levels that Showed Highest Differences Due to Tested Tissues. Blue color shows well-watered traits; red color represents drought-stressed traits. Bright colors highlight stress-susceptible traits; dark colors accent tolerant traits. Each value represents the mean and SD of six individual biological replicates.

CONCLUSIONS The greenhouse drought-stress experiment conducted in Irapuato (Guanajuato, MX) in 2009 give new possibilities into investigating drought stress in maize combining physiological and metabolic datasets. Six different hybrids were analyzed and, in all traits, differences in plant height, transpiration, ear height, leaf temperature, and chlorophyll content due to the drought-stress treatment was monitored. Plant height was affected from drought in all hybrids. Furthermore, leaf temperature increased due to the water limitation in all hybrids. An increase in ear height could be observed in hybrids

NPE1, NPE2, NPE3, and NPE5. A decrease in chlorophyll content was detected for all lines except NPE4. However, significant differences regarding the predicted drought tolerance could not be verified. Thus, the metabolic profile analyses also did not show differences in between these three tolerance groups (susceptible, moderate, and tolerant). Selection of described genotypes and its predicted drought tolerance were based on field trial data that could not be confirmed in this greenhouse experiment. The fact that a natural setting cannot be fully transferred to a greenhouse may be a possible reason for this. Nevertheless, the metabolic pattern of all tested traits showed a large number of metabolites that were significantly

Witt et al.

d


|

411

Figure 8. Box Plots of Selected Metabolite Levels that Showed Highest Differences Due to the Drought Treatment in Tested Tissues. Blue color shows well-watered traits; red color represents drought-stressed traits. Bright colors highlight stress-susceptible traits; dark colors accent tolerant traits. Each value represents the mean and SD of six individual biological replicates.

changed due to the applied treatment. Analyses of different maize tissues like leaf blade, leaf sheath, ear, husk, and silks showed that the most contrasting metabolic pattern due to drought-stress treatment was observed in leaf blade. At the top with highest p-values regarding the treatment effect are amino acids like tryptophan, proline, and histidine, sugars like fucose, and several intermediates from the TCA cycle. Focusing on leaf blade tissue, we found several compounds that showed significant differences between tested hybrids. Summarizing evaluated data, we could verify a strong correlation between phenotypical changes in maize during drought-stress response or/and adaptation and changes in plant metabolism.

When compared with the data from wheat presented in this issue (Bowne et al., 2012), similar findings were found, suggesting that this information will likely be useful to find more tolerant crop genotypes in the future and will improve plant breeding.

METHODS Plant Material, Growth Conditions, and Experimental Design Seeds from six different maize hybrids were planted under well-watered and drought-stress conditions in the greenhouse in Irapuato (Guanajuato, Mexico) in 2009 (Table 1). Selected

412

|

Witt et al.

d


profiling. Two plants per genotype were grown into physiological maturity.

GC–TOF–MS Analysis of Metabolite Abundance

Figure 9. Box Plots of Selected Metabolite Levels that Showed Highest Differences Due to Predicted Drought Tolerance in Tested Tissues. Blue color shows well-watered traits; red color represents droughtstressed traits. Bright colors highlight stress-susceptible traits; dark colors accent tolerant traits. Each value represents the mean and SD of six individual biological replicates.

hybrids were part of the germplasm drought breeding program at CIMMYT. A prediction of drought tolerance of the tested traits was made based on nine independent field experiments. For experimental conditions of the field experiments, see Monneveux et al. (2008). The presented data represent results of six biological replicates of six different hybrids for every measured parameter. In the greenhouse, one plant per 10-L pot was planted in 1:1 peatmoss and vermiculite mixed soil and optimal fertilized. Water was controlled by drip irrigation with 2 L water per hour per pot. Two weeks before flowering, drought stress was applied by stopping irrigation for 12 d. Stomatal conductance, chlorophyll content, plant height, ear height, and leaf temperature were monitored to assess the stress level. In vivo chlorophyll content of leaves was measured using a portable chlorophyll meter (SPAD-502, Minolta, Tokyo, Japan). Plant water status was estimated by measuring leaf temperature and stomatal conductance on the abaxial surface of sun-exposed leaves from the upper part of the plant with a Decagon Leaf Porometer SC-1 (Decagon Device Inc., Pullman, WA, USA). Eight weeks after planting, the leaf blade, ear, and husks were harvested. All samples were snap-frozen in liquid nitrogen, lyophilized, and sent to Potsdam-Golm for metabolic

Metabolites for GC–TOF–MS were extracted using a protocol adapted from Lisec et al. (2006) and Roessner et al. (2001). After sampling, the material was freeze-dried for 1 week in a desiccator. For the extraction, 100 mg of lyophilized flour from different tissue was used and mixed with 60 ll Ribitol (0.2 mg ml 1 stock in water) as an internal quantitative standard for the polar phase. After shaking the mix for 15 min at 70C, the extract was centrifuged for 10 min at 14 000 rpm. For separation of polar and non-polar metabolites, the supernatant was very carefully mixed with 750 ll Chloroform and 1500 ll water. After centrifugation for 15 min at 4000 rpm, polar and non-polar fraction was separated and a 150-ll aliquot from the upper polar phase was taken off and dried in vacuo. Samples were then shipped to Golm, Potsdam, for metabolic profile. The pellet was re-suspended in 60 ll methoxyaminhydrochloride (30 mg ml 1 in pyridine) and shaken for 2 h at 37C. After this, 1 ml of MSTFA (N-methyl-N-[trimethylsilyl] trifluoroacetamide) containing 20 ll retention timestandard mixture of fatty acid methylesters (methylcaprylate, methyl pelargonate, methylcaprate, methyllaurate, methylmyristate, methylpalmitate, methylstearate, methyleicosanoate, methyldocosanoate, lignoceric acid methylester, methylhexacosanoate, methyloctacosanoate, triacontanoic acid methylester, d6-cholesterol) was added. This mix was incubated for 30 min at 37C. 1 ll of each sample were used to inject into a GC–TOF–MS system (Pegasus III, Leco, St Joseph, USA). Gas chromatography was performed using a 30-m MDN-35 column. The injection temperature was 230C whereas the transfer line and ion source were set at 250C. The initial oven temperature of 85C was constantly increased to a final temperature of 360C by increments of 15C per minute. Mass spectra were recorded at 20 scans per second with an m/z 70–600 scanning range. Chromatograms and mass spectra were evaluated and metabolite levels determined in a targeted fashion using a library derived from the GolmMetabolome-Database (GMD; Kopka et al., 2005). Each metabolite is represented by the observed ion intensity of a selected unique ion that allows for a relative quantification between groups. Metabolite data were log10-transformed to improve normality and normalized to show identical medium peak sizes per sample group (Steinfath et al., 2008).

Statistical Analysis Statistical analyses and graphical representations (Student’s t-test, Analysis of Variance (ANOVA), Bonferroni-correction, principal component analysis (PCA), boxplots) were performed using the R-software environment 2.13.0 (http://cran.r-project. org/). The PCA was conducted using the ‘bpca’ algorithm of the pcaMethods package (Stacklies et al., 2007). ANOVA was conducted using Genotype, Treatment, and Tolerance as factors. Resulting P-values were corrected for multiple testing using the stringent Bonferroni method.

Witt et al.

d


|

413

Figure 10. Heatmap of Metabolite Levels in Different Tissues and Genotypes Normalized to Well-Watered Conditions. Red colors represent increase in metabolite levels; blue colors highlight decreases using a false-color scale. Values represent the mean of each genotype (n = 6).

414

|

Witt et al.

d


Table 3. Two-Way ANOVA of the Metabolic Data.

Table 3. Continued

Analyte blade

Analyte blade

Treatment p–value

Treatment p–value

Tryptophan

1.09E-17

Tyramine

9.06E-03

Histidine

2.01E-16

Alanine, beta-

9.35E-13

Butanoic acid, 2-amino-

6.71E-16

2-Piperidinecarboxylic acid

7.14E-10

Phenylalanine

2.07E-15

Spermidine

1.58E-07

Putrescine

5.23E-14

Galactinol

2.14E-05

Glycine

1.19E-13

Fucose

2.24E-05

Methionine

1.50E-12

Ferulic acid, trans-

4.07E-05

Asparagine

6.11E-12

Caffeic acid, trans-

7.06E-05

Tyrosine

7.44E-12

Glucopyranose

1.50E-04

Galactinol

2.82E-11

Propylamine-2,3-diol

1.52E-04

Serine

2.96E-11

Androst-5-en-17-one, 3beta-hydroxy-

1.61E-04

Alanine, beta-

3.26E-11

Itaconic acid

2.12E-04

Glutamine

3.55E-11

Alanine

2.97E-04

Isoleucine

5.78E-11

Serine, O-acetyl-

4.71E-04

Pyroglutamic acid

5.94E-11

Caffeic acid, cis-

4.76E-04

Quinic acid, 3-caffeoyl-, trans-

1.97E-10

Kestose, 1-

4.80E-04

Dihydrosphingosine

2.91E-09


6.74E-04

Quinic acid, 3-caffeoyl-, cis-

3.50E-09

Phosphoric acid

8.18E-04

Proline

1.21E-08

Quinic acid, 3-caffeoyl-, cis-

9.17E-04

Adenine

2.55E-08

Fumaric acid, 2-methyl-

1.26E-03

Pyruvic acid

9.33E-08

Glycerol

1.96E-03

2-Piperidinecarboxylic acid

2.40E-07

Rhamnose

1.97E-03

Succinic acid

3.23E-07

myo-Inositol-1-phosphate

2.25E-03


6.23E-07

Glycerol-3-phosphate

2.27E-03

Valine

1.92E-06

Erythritol

3.03E-03

Fucose

3.37E-06

Dihydrosphingosine

3.16E-03

Turanose

6.24E-06

Nicotinamide

4.32E-03

Ornithine

1.25E-05

Turanose

5.23E-03

Glutaric acid, 2-oxo-

1.67E-05

Cinnamic acid, 4-hydroxy-, trans-

7.12E-03

Kestose, 1-

4.80E-05

Itaconic acid

1.39E-05

Cytidine-2’,3’-cyclic-monophosphate

5.07E-05

Caffeic acid, trans-

1.45E-05

Homoserine

5.32E-05

Fumaric acid, 2-methyl-

5.81E-05

Jasmonic acid methyl ester, 2-trans-

8.28E-05

Serine, O-acetyl-

7.39E-05

Xylose

8.61E-05

Caffeic acid, cis-

1.35E-04

Glucosamine, N-acetyl-

1.13E-04

Glucopyranose

1.62E-04

Propylamine-2,3-diol

1.15E-04

Galactinol

1.89E-04

Nicotinamide

4.57E-04

Phosphoric acid

5.11E-04


4.72E-04

Spermidine

1.04E-03

Urea

6.75E-04

Cinnamic acid, 4-hydroxy-, trans-

1.49E-03

Lactulose

1.39E-03

Hexadecanoic acid, 3-hydroxy-

1.64E-03

Glucopyranose

1.43E-03

Fucose

4.57E-03

Leucrose

3.05E-03

myo-Inositol-1-phosphate

5.07E-03

Idose

3.91E-03


6.83E-03

Sorbose

5.62E-03

Alanine, beta-

8.30E-03

Aconitic acid, trans-

7.25E-03

Glutamic acid

8.74E-03

Erythritol

9.03E-03

Lists for treatment, genotype, and tolerance classification are provided. Gray squares represent significant association (P , 0.01).

Witt et al.

d


|

415

Figure 11. Bonferroni-Corrected Correlation Analysis. Blue color represents well-watered samples; red color represents drought-stressed samples. Bright colors highlight stress-susceptible genotypes; dark colors accent tolerant genotypes. Mean values are shown. (A) Relation of leaf temperature after 7 d of water restriction and abundance of pyruvic acid and succinic acid under well-watered and water-restricted conditions is shown as well as for the amino acids isoleucine and proline. (B) Relation of leaf stomatal conduction after 1 d of water restriction and abundance of TCA cycle intermediates pyruvic acid and succinic acid under well-watered and water-restricted conditions is shown as well as for the amino acids isoleucine and proline. (C) Relation of leaf stomatal conductance after 7 d of water restriction and abundance of TCA cycle intermediate pyruvic acid, amino acids phenylalanine, alanine and the organic compound putrescine.

Supplementary Data are available at Molecular Plant Online.

fu¨r Wirtschaftliche Zusammenarbeit und Entwicklung (BMZ), Germany, and in part by the OPTIMAS project of the Bundesministerium fu¨r Bildung und Forschung (BMBF) Germany.

FUNDING

ACKNOWLEDGMENTS

This work was supported in part by the ‘Precision phenotyping for improving drought-stress tolerant maize in southern Asia and eastern Africa’ project, funded by the Bundesministerium

We are additionally grateful to Ciro Sanchez and the personnel of the maize quality lab at CIMMYT for their technical support. No conflict of interest declared.

SUPPLEMENTARY DATA

416

|

Witt et al.

d


REFERENCES Alcazar, R., Bitrian, M., Bartels, D., Koncz, C., Altabella, T., and Tiburcio, A.F. (2011). Polyamine metabolic canalization in response to drought stress in Arabidopsis and the resurrection plant Craterostigma plantagineum. Plant Signaling & Behavior. 6, 243–250. Araus, J.L., Brown, H.R., Febrero, A., Bort, J., and Serret, M.D. (1993). Ear photosynthesis, carbon isotope discrimination and the contribution of respiratory CO2 to differences in grain mass in durum wheat. Plant Cell and Environment. 16, 383–392. Araus, J.L., Slafer, G.A., Reynolds, M.P., and Royo, C. (2002). Plant breeding and drought in C3 cereals: what should we breed for? Ann. Bot. 89, 925–940. Ba¨nziger, M., Edmeades, G.O., and Lafitte, H.R. (2002). Physiological mechanisms contributing to the increased N stress tolerance of tropical maize selected for drought tolerance. Field Crops Res. 75, 223–233. Bartels, D., and Nelson, D. (1994). Approaches to improve stress tolerance using molecular genetics. Plant Cell Environ. 17, 659–667. Bartels, D., and Sunkar, R. (2005). Drought and stress tolerance in plants. Crit. Rev. Plant Sci. 24, 23–58. Bohnert, H.J., Gong, Q., Li, P., and Ma, S. (2006). Unravelling abiotic stress tolerance mechanisms: getting genomics going. Curr. Opin. Plant Biol. 9, 180–188. Bolanos, J., and Edmeades, G.O. (1996). The importance of the anthesis-silking interval in breeding for drought tolerance in tropical maize. Field Crop Res. 48, 65–80. Bowne, J.B., Erwin, T.A., Juttner, J., Schnurbusch, T., Langridge, P., Bacic, A., and Roessner, U. (2012). Drought responses of leaf tissues from wheat cultivars of differing drought tolerance at the metabolite level. Mol. Plant. 5, 418–429. Bruce, W.B., Edmeades, G.O., and Barker, T.C. (2002). Molecular and physiological approaches to maize improvement for drought tolerance. J. Exp. Bot. 53, 13–25. Cabrera-Bosquet, L., Sanchez, C., and Araus, J.L. (2009). How yield relates to ash content, d13C and d18O in maize grown under different water regimes. Ann. Bot. 104, 1207–1216. Cabrera-Bosquet, L., Sanchez, C., Rosales, A., Palacios-Rojas, N., and Araus, J.L. (2011). Near-infrared reflective spectroscopy (NIRS) assessment of d18O and nitrogen and ash contents for improved yield potential and drought adaptation in maize. J. Agri. Food Chem. 59, 467–474. Chourey, P.S., Li, Q.B., and Kumar, D. (2010). Sugar-hormone crosstalk in seed development: two redundant pathways of IAA biosynthesis are regulated differentially in the invertase-deficient minature1 (mn1) seed mutant in maize. Mol. Plant. 3, 1026–1036. Crossa, J., et al. (2010). Prediction of genetic values of quantitative traits in plant breeding using pedigree and molecular markers. Genetics. 186, 713–724. Edmeades, G.O., Ba¨nziger, M., and Cortez, C. (1996a). From stresstolerant populations to hybrids: the role of source germ plasm. In Developing Drought and Low N-Tolerant Maize, Edmeades, G.O., Ba¨nziger, M., Mickelson, H.R., and Penã-Valdivia, C.B., eds, proceedings of a symposium, 25–29 March 1996, CIMMYT, El Batań, Mexico (Mexico DF: CIMMYT), pp. 274–284.

Edmeades, G.O., Bolanos, H.R., and Chapman, S.C. (1996b). Value of secondary traits in selecting for drought tolerance in tropical maize. In Developing Drought and Low N-Tolerant Maize, Edmeades, G.O., Ba¨nziger, M., Mickelson, H.R., Penã-Valdivia, C.B., eds, proceedings of a symposium, 25–29 March 1996, CIMMYT, El Batań, Mexico (Mexico DF: CIMMYT), pp. 233–245. Edmeades, G.O., Bolanos, H.R., Lafitte, S., Rajram, S., Pfeiffer, W., and Fischer, R.A. (1989). Traditional approaches to breeding for drought resistance in cereals. In Drought Resistance in Cereals, Baker, F.W.G., ed. (Wallingford, UK: ICSU and CABI), pp. 27–52. Edmeades, G.O., Bolanos, J., Chapman, S.C., Lafitte, H.R., and Ba¨nziger, M. (1999). Selection improves drought tolerance in tropical maize populations. I. Gains in biomass, grain yield, and harvest index. Crop Science. 39, 1306–1315. Elings, A., White, J., and Edmeades, G.O. (1996). Modeling the consequences of water limitations at flowering in tropical maize germplasm. In Developing Drought and Low N-Tolerant Maize, Edmeades, G.O., Ba¨nziger, M., Mickelson, H.R., Penã-Valdivia, C.B., eds, proceedings of a symposium, 25–29 March 1996, CIMMYT, El Batań, Mexico (Mexico DF: CIMMYT), pp. 151–158. Fernie, A.R., and Klee, H.J. (2011). The use of natural genetic diversity in the understanding of metabolic organization and regulation. Front Plant Sci. 2, 59. Fernie, A.R., and Schauer, N. (2009). Metabolomics-assisted breeding: a viable option for crop improvement? Trends Genet. 25, 39–48. Fernie, A.R., Tadmor, Y., and Zamir, D. (2006). Natural genetic variation for improving crop quality. Curr. Opin. Plant Biol. 9, 196–202. Harjes, C.E., et al. (2008). Natural genetic variation in lycopene e cyclase tapped for maize biofortification. Science. 319, 330–333. Harrigan, G.G., et al. (2007a). Impact of genetics and environment on nutritional and metabolite components of maize grain. J. Agri. Food Chem. 55, 6177–6185. Harrigan, G.G., et al. (2007b). Metabolite analysis of grain from maize hybrids grown under drought and watered conditions during the 2002 field season. J. Agri. Food Chem. 55, 6169–6176. Hoisington, D., Jiang, C., Khairallah, M., Ribaut, J.M., Bohn, M., Melchinger, A., Wilcox, M., and Gonzalez-de-Leon, D. (1996). QTL for insect resistance and drought tolerance in tropical maize: prospects for marker assisted selection. Symposium Soc. Exp. Biol. 50, 39–44. Kochevenko, A., Arau´jo, W.L., Maloney, G.S., Tieman, D.M., Do, P.T., Taylor, M.G., Klee, H.J., and Fernie, A.R. (2012). Catabolism of branched chain amino acids supports respiration but not volatile synthesis in tomato fruits. Mol. Plant. 5, 366–375. Kopka, J., et al. (2005). [email protected]: the Golm Metabolome Database. Bioinformatics. 21, 1635–1638. Leung, J., and Giraudat, J. (1998). Abscisic acid signal transduction. Annu. Rev. Plant Physiol. Plant Mol. Biol. 49, 199–222. Lisec, J., Ro¨misch-Margl, L., Nikoloski, Z., Piepho, H.P., Giavalisco, P., Selbig, J., Gierl, A., and Willmitzer, L. (2011). Corn hybrids display lower metabolite variability and complex metabolite inheritance patterns. Plant J. doi: 10.1111/j.1365–313X.2011.04689.x. Lisec, J., Schauer, N., Kopka, J., Willmitzer, L., and Fernie, A.R. (2006). Gas chromatography mass spectrometry-based metabolite profiling in plants. Nat. Protoc. 1, 387–396.

Witt et al.

Lopes, M.S., Araus, J.L., Van Heerden, P.D.R., and Foyer, C.H. (2011). Enhancing drought tolerance in C4 crops. J. Exp. Bot. 62, 3135–3153. Monneveux, P., Sanchez, C., and Tiessen, A. (2008). Future progress in drought tolerance in maize needs new secondary traits and cross combinations. J. Agri. Sci. 146, 1–14. Monneveux, P., Sanchez, C., Beck, D., and Edmeades, G.O. (2005). Drought tolerance improvement in tropical maize source populations: evidence of progress. Crop Sci. 46, 180–191. Moose, S.P., et al. (2004). Maize selection passes the century mark: a unique resource for 21st century genomics. Trends Plant Sci. 9, 358–364. Nunes-Nesi, A., Fernie, A.R., and Stitt, M. (2010). Metabolic and signaling aspects underpinning the regulation of plant carbon nitrogen interactions. Mol. Plant. 3, 973–996. Pandey, S., and Gardner, C.O. (1992). Recurrent selection for population, variety, and hybrid improvement in tropical maize. Adv. Agron. 48, 1–87. Quarrie, S.A., Lazic-Jancic, V., Kovacevic, D., Steed, A., and Pekic, S. (1999). Bulk segregant analysis with molecular markers and its use for improving drought resistance in maize. J. Exp. Bot. 50, 1299–1306. Ristic, Z., and Cass, D.D. (1981). Leaf anatomy of Zea mays L. in response to water shortage and high temperature: a comparison of drought resistant and drought-sensitive lines. Botanical Gazette. 152, 173–185. Roessner, U., Luedemann, A., Brust, D., Fiehn, O., Linke, T., Willmitzer, L., and Fernie, A.R. (2001). Metabolic profiling allows comprehensive phenotyping of genetically or environmentally modified plant systems. Plant Cell. 13, 11–29. Ro¨misch-Margl, L., Spielbauer, G., Schu¨tzenmeister, A., Schwab, W., Piepho, H.P., Genschel, U., and Gierl, A. (2010). Heterotic patterns of sugar and amino acid components in developing maize kernels. Theo. Appl. Genet. 120, 369–381. Rontein, D., Basset, G., and Hanson, A.D. (2002). Metabolic engineering of osmoprotectant accumulation in plants. Met. Eng. 4, 49–56.

d


|

417

and non-irrigated field grown tomato of varying genotype. Metabolomics. 3, 289–295. Setter, T.M., Yan, J., Warburton, M., Ribaut, J.M., Xu, Y., Sawkins, M., Buckler, E.S., Zhang, Z., and Gore, M.A. (2011). Genetic association mapping identifies single nucleotide polymorphisms in genes that affect abscisic acid levels in maize floral tissues during drought. J. Exp. Bot. 62, 701–716. Sharma, S., Villamor, J.G., and Verslues, P.E. (2011). Essential role of tissue-specific proline synthesis and catabolism in growth and redox balance at low water potential. Plant Physiol. 157, 292–304. Stacklies, W., Redestig, H., Scholz, M., Walther, D., and Selbig, J. (2007). pcaMethods: a bioconductor package providing PCA methods for incomplete data. Bioinformatics. 23, 1164–1167. Steinfath, M., Groth, D., Lisec, J., and Selbig, J. (2008). Metabolite profile analysis: from raw data to regression and classification. Physiol. Plant. 132, 150–161. Sun, Y., Helentjaris, T., Zinselmeier, C., and Habben, J.E. (2001). Utilizing gene expression profiles to investigate maize response to drought stress. In 1999 Proceedings of the 54th Annual Corn and Sorghum Research Conference. (Washington DC: American Seed Trade Association ), pp. 140–153. Tambussi, E.A., Bort, J., Nogue´s, S., Guiamet, J.J., and Araus, J.L. (2007). The photosynthetic role of ears in C3 cereals: metabolism, water use efficiency and contribution to grain yield. Critical Reviews on Plant Science. 26, 1–16. Tardieu, F., Granier, C., and Muller, B. (2011). Water deficit and growth: co-ordinating processes without an orchestrator? Curr. Opin. Plant Biol. 14, 283–289. Tollenaar, M., and Wu, J. (1999). Yield improvement in temperate maize is attributable to greater stress tolerance. Crop Sci. 39, 1597–1604. Verbruggen, N., and Hermans, C. (2008). Proline accumulation in plants: a review. Amino Acids. 35, 753–759. Xu, Y., Skinner, D.J., Wu, H., Palacios-Rojas, N., Araus, J.L., Yan, J., Gao, S., Warburton, M.L., and Crouch, J.H. (2009). Advances in maize genomics and their value for enhancing genetic gains from breeding. Int. J. Plant Genomics. 2009, 957602.

Rosen, S., and Scott, L. (1992). Famine grips sub-Saharan Africa. Outlook Agric. 191, 20–24.

Yancey, P.H. (2005). Organic osmolytes as compatible, metabolic and counteracting cytoprotectants in high osmolarity and other stresses. J. Exp. Biol. 208, 2819–2830.

Schauer, N., et al. (2006). Comprehensive metabolic profiling and phenotyping of interspecific introgression lines for tomato improvement. Nature Biotech. 24, 447–454.

Yang, S., Vanderbeld, B., Wan, J., and Huang, Y. (2010). Narrowing down the targets: towards successful genetic engineering of drought-tolerant crops. Mol. Plant. 3, 469–490.

Schauer, N., Semel, Y., Balbo, I., Steinfath, M., Repsilber, D., Selbig, J., Pleban, T., Zamir, D., and Fernie, A.R. (2008). Mode of inheritance of primary metabolic traits in tomato. Plant Cell. 20, 509–523.

Zheng, P., et al. (2008). A phenylalanine in DGAT is a key determinant of oil content and composition in maize. Nat. Genet. 40, 367–372.

Seki, M., Narusaka, M., Abe, H., Kasuga, M., YamaguchiShinozaki, K., Carninci, P., Hayashizaki, Y., and Shinozaki, K. (2001). Monitoring the expression pattern of 1300 Arabidopsis genes under drought and cold stresses by using a full-length cDNA microarray. Plant Cell. 13, 61–72. Semel, Y., Schauer, N., Roessner, U., Zamir, D., and Fernie, A.R. (2007). Metabolite analysis for the comparison of irrigated

Zinselmeier, C., Habben, J.E., Westgate, M.E., and Boyer, J.S. (2000). Carbohydrate metabolism in setting and aborting maize ovaries. In Physiology and Modeling Kernel Set in Maize, CSSA Special publication number 29, Westgate, M.E., Boote, K., eds (Madison: Crop Science Society of America), pp. 1–14. Zinselmeier, C., Lauer, M.J., and Boyer, J.S. (1995). Reversing drought-induced losses in grain yield: sucrose maintains embryo growth in maize. Crop Sci. 35, 1390–1400.